Legal & DebateNEUTRAL — Rhetoric

Modus Tollens Refutation

What it is

Refuting a claim through its predictions: if the claim were true, some consequence would be observed; the consequence is not observed; therefore the claim is false — the logical form behind falsification and most honest empirical rebuttal.

How it works

Modus tollens is the valid inference from “if P then Q” and “not Q” to “not P.” It is the engine of Karl Popper's account of science: a theory that says all swans are white is not confirmed by any number of white swans but is refuted by one black one, and a theory that forbids nothing predicts nothing. In argument the form turns a disagreement into a test — “if your explanation were right, we would see X; do we?” — and it is the most useful move for exposing unfalsifiable claims, which produce no Q that could fail to appear. Two features complicate it. First, people find the form hard: Peter Wason's selection task and decades of follow-up show that reasoners endorse modus ponens almost universally but modus tollens far less often, and confuse it with the invalid form of denying the antecedent. Second, Pierre Duhem observed that predictions follow not from a theory alone but from the theory plus auxiliary assumptions, so a failed prediction can always be blamed on an auxiliary; the honest reasoner limits such rescues, the motivated one multiplies them. The manipulative use is a fake conditional — “if you cared, you would have called” — where the “if” was never true and the missing call now proves the missing care.

Real-world examples

  • •Popper's black swan: “all swans are white” forbids black swans; Dutch sailors' sighting of black swans in Western Australia in 1697 ended the generalization for good.
  • •Debugging: “If the fault were in the cache, clearing it would fix the page. It did not. The fault is elsewhere.” Every engineer runs modus tollens daily and rarely names it.
  • •A debate rebuttal: “If the plan really deterred, states facing it would have changed behavior; their own evidence shows none did; the deterrence claim fails.”
  • •Duhem's escape in the wild: a forecaster whose prediction failed explains that it would have succeeded but for an unforeseen shock — sometimes true, and unfalsifiable if invoked every time.
  • •Family: “If you respected me you would have asked first; you did not ask; so you do not respect me” — valid in form, and unsound if asking first was never what respect required.

Ethical guidelines

  • ●State the conditional before the observation, not after; a prediction announced once the result is known is not a test.
  • ●When your prediction fails, say what you will revise rather than reaching for an auxiliary assumption; count your rescues.
  • ●Do not build conditionals the other person never accepted in order to convict them on the consequent.
  • ●Accept a modus tollens run on you: if your view predicted something that did not happen, your view has a problem, whoever points it out.

How to defend against it

  • ►Ask for the prediction in advance: “What would we expect to see if you are right, and what would we see if you are wrong?” A claim that has the same answer to both is not a claim.
  • ►Challenge the conditional, not the observation: “Yes, I did not call. No, calling was not what caring required. Your first premise is false.” Most manipulative modus tollens fails at the “if.”
  • ►When a failed prediction is blamed on an auxiliary, ask whether the auxiliary was stated before the result and whether it would have been invoked had the prediction succeeded.
  • ►Check for the invalid cousin: “if P then Q; not P; so not Q” proves nothing. “If the policy worked, crime would fall; the policy was not tried; so crime will not fall” is denying the antecedent.

References

  1. Popper, K. R. (1959). The Logic of Scientific Discovery. Hutchinson (original German edition, Logik der Forschung, 1934)
    Falsification as the logic of empirical testing, with modus tollens as its inferential form.
  2. Wason, P. C. (1968). Reasoning about a Rule. Quarterly Journal of Experimental Psychology, 20(3), 273-281
    Experimental evidence that people reason poorly with conditional rules, including modus tollens.
  3. Evans, J. St. B. T., Newstead, S. E., & Byrne, R. M. J. (1993). Human Reasoning: The Psychology of Deduction. Lawrence Erlbaum
    Review showing lower endorsement of modus tollens than modus ponens across studies.
  4. Duhem, P. (1954). The Aim and Structure of Physical Theory (trans. P. P. Wiener; original French 1906). Princeton University Press
    Predictions depend on auxiliary hypotheses, so a failed test can be deflected onto an auxiliary.
Last reviewed
Suggest a correction

Detect Modus Tollens Refutation in any text

Paste any message, email, or article into our free Manipulation Detector to see if Modus Tollens Refutation or other techniques are being used on you.

Try the Detector FreeCreate Free Account